% latex table generated in R 4.1.2 by xtable 1.8-4 package
% Sun Feb 13 22:04:47 2022
\begingroup\tiny
\begin{longtable}{p{0.07\textwidth}p{0.05\textwidth}p{0.1\textwidth}p{0.4\textwidth}p{0.1\textwidth}p{0.1\textwidth}p{0.2\textwidth}}
\caption{Literature Review} \\ 
  \toprule
Study & Time frame & Subjects & Key findings (as described by the authors) & Subgroups & DVs tested & Concerns \\ 
  \midrule
\cite{warner1991does} & 1988--1999 & Toronto and Detroit & Offspring sex does not affect American men's views on gender, but American men and women and Canadian women with only female children hold significantly more feminist views than those with only male children. Canadian men and women with only female children are not significantly more feminist than those with female and male children but American women with only female children are significantly more feminist than those with female and male children. & men/women, Canada/US
 & Feminist orientation (six survey items taken from Carroll (1985)) & Specification: compares any daughter(s)/any son(s) to no children, effect of no. of children may be nonlinear, analysis of US women: combines sons only and both sons and daughters as new variable (= any male children)
Inclusion of nonbiological offspring: ambiguous
Post-treat: controls for ideology \\ 
   \midrule
\cite{katzev1994girls} & 1987--1988 & US & Mothers of only boys are more likely to believe in a more nontraditional family ideology than mothers of only girls or mothers with children of both sexes. But mothers of boys and girls are similar to those with only girls. & fathers/mothers & Marital instability (likelihood of seperation, nontraditional family ideology, perceptions of equity, time with children, time in housework) & Specification: no measure of proportion of daughters, effect of no. of children may be nonlinear
Inclusion of nonbiological offspring: yes (controls for presence of adopted/step-children)
Post-treat: controls for nontraditional family ideology \\ 
   \midrule
\cite{downey1994sons} & 1990 & Indiana mothers & Mothers with more sons than daughters (or with a larger proportion of sons) have more negative views regarding maternal employment and voice greater concern over the need for obedience by children. &  & Level of agreement with the following statements: 'Children always suffer when both parents work outside the home' and 'The most important thing for children to learn is to obey' & Specification: effect of no. of children may be nonlinear
Inclusion of nonbiological offspring: ambiguous \\ 
   \midrule
\cite{warner1999child} & 1997--1998 & Oregon and Washington & Parents of only daughters support feminist policies more than parents of only sons or parents with no children. The effect is larger for fathers than for mothers. For both women and men, the highest levels of support come from those with daughters only, whereas the lowest levels come from those with sons only. However, for women, the significant comparison is for daughters only compared to other categories. Men who have no children or have both sons and daughters hold similar opinions about gender equity public policies. & men/women & Support for gender equity policies & Specification: no measure of proportion of daughters, effect of no. of children may not be linear
Inclusion of nonbiological offspring: ambiguous
Post-treat: controls for gender role attitudes, political ideology \\ 
   \midrule
\cite{washington2008female} & 1997--2005 & US Representatives & Each daughter increases a congressperson's propensity to vote liberally on women's issues by about two points, which is about 25 percent of the
size of the own gender ``effect.'' & men/women, Democrats/Republicans & NOW, AAUW, and NRLC scores & Inclusion of nonbiological offspring: yes
Post-treat: controls for Representatives' partisan affiliation \\ 
   \midrule
\cite{oswald2010daughters} & 1991--2005 & British and German adults & For each daughter, holding family size constant, a parent is approximately two percentage points more likely to vote left. & men/women & Partisan identification, attitudes on traditional gender roles & Inclusion of nonbiological children: in some analyses \\ 
   \midrule
\cite{prokos2010attitudes} & 2000, 2002, 2004, and 2006 & US employed population & Mothers with only daughters are more likely to support affirmative action for women than mothers with only sons or children of either gender. The reverse is true for fathers---fathers with only daughters are less likely to support affirmative action for women than commensurate groups. Women and men with no children, a child of each gender, and sons only are equally likely to support gender-based affirmative action policies. & men/women & Support for gender-based affirmative action (special efforts to hire and promote women) & Specification: no measure of proportion of daughters, uses categorical specification without controlling for no. of children
Inclusion of nonbiological offspring: yes
Post-treat: controls for traditional gender role attitudes, belief that discrimination does not exist, conservative political views \\ 
   \midrule
\cite{shafer2011effect} & 1979, 1982, 1987 and 2004 & US adults & Having a daughter reduces support for traditional gender roles among men, but the effect size is small---.1 sd. Among women, there is no effect of having a daughter---.04 sd.
 & fathers/mothers & Support for traditional gender roles, within-subject attitudinal shifts (longitudinal data) & Inclusion of nonbiological offspring: yes \\ 
   \midrule
\cite{iacus2011multivariate} & 1997--2005 & US Representatives & Replication of Washington (2008). & Apart from use of matching, same design as Washington (2008) &  &  \\ 
   \midrule
\cite{conley2013effect} & 1994 & US adults & An increase in the number of girls from none to one translates on average into a 9\% increase in the proportion of gender-related cases in which a judge will vote in a feminist direction. Having at least one daughter corresponds to a 7\% increase in the proportion of cases in which a judge will vote in a feminist direction. Effect for Republican judges is an average 7\% increase, which is significant. Effect for Democrat judges is on average 4% and is not significant. The greatest change comes simply from having at least one girl; there is no added impact on having additional girls
 & men/women, low/high social economic status & Traditional views of women, partisan affiliation, pro-abortion index, attitudes on teen sex & Inclusion of nonbiological offspring: ambiguous
Other data concerns: small sample size \\ 
   \midrule
\cite{glynn2015identifying} & 1996-2002 & US Courts of Appeals judges & An increase in the number of girls from none to one translates on average into a 9% increase in the proportion of gender-related cases in which a judge will vote in a feminist direction. Having at least one daughter corresponds to a 7% increase in the proportion of cases in which a judge will vote in a feminist direction. Effect for Republican judges is an average 7% increase, which is significant. Effect for Democrat judges is on average 4% and is not significant. The greatest change comes simply from having at least one girl; there is no added impact on having additional girls
 & men/women, Democrat/Republican, judges with 1-4 children, judges with 1 child & Case-level Judge-vote in feminist direction (gender-related cases)
Case-level Judge-vote in liberal direction (all cases) & Post-treat: controls for Judges' partisan affiliation \\ 
   \midrule
\cite{lee2016does} & 1972-2012 (US)/2002-2012 (EU) & US/European adults & In the UK, if parents have a girl for their first child, they are more likely to lean toward the Conservative Party, though the effects are insignificant.  In the US, first daughter has a significant conservative effect in 1994 and significant liberal effect in 2002 and 2004.  Authors conclude null effects of the sex of the first child on party identification as well as on political ideology while ruling out country heterogeneity. & over time, by country (Table 2 restricts focus to US and UK) & Partisan affiliation, political ideology & Inclusion of nonbiological offspring: yes
Other data concerns: authors exclude 90\% of cases from GSS, 80\% of cases from ESS. In both GSS and ESS, exclusions cannot be considered random in any way (i.e. authors restrict their sample to parents whose oldest child is younger than seventeen, thus their sample is biased towards younger families (and, as Figure 1 indicates, smaller families)) \\ 
   \midrule
\cite{sun2017mothers} & 2012--2013 & Chinese parents & Having sons causes mothers to develop significantly more traditional gender ideologies than fathers (but the effect of having sons on traditional gender ideology among all parents is statistically insignificant). Compared with parents who had only daughters, parents who had only sons or had both sons and daughters did not show significant difference in gender ideology. Mothers with both sons and daughters had more traditional gender ideologies than mothers with only daughters. In 2012 sample, the interaction term of motherhood and having only sons showed a significant positive correlation with traditional gender ideology. & fathers/mothers & Gender ideology (level of agreement with five statements) & Specification: no measure of proportion of daughters, effect of no. of children may be nonlinear
Inclusion of nonbiological offspring: yes
Post-treat: controls for Communist party membership \\ 
   \midrule
\cite{yu2018does} & 2007, 2009, 2011 & Japanese adults & Japanese men adopted a less traditional view, acknowledging that having children hinders a couple�s life, upon the arrival of a son, but not a daughter. Having an additional son is equally likely as having an additional daughter to be associated with lowered support for the notion that maternal employment is harmful for young children among Japanese women and men. Adding a daughter (but not a son) reduces Japanese women�s disapproval of divorce. & men/women & Attitudes towards gender and family issues (including traditional gender division of labor) and within-suubject attitudinal shifts & Inclusion of nonbiological offspring: ambiguous (unclear whether JLPS dataset includes nonbiological children) \\ 
   \midrule
\cite{sharrow2018first} & 2016 & US parents & Having a daughter first positively influences attitudes toward sex-equity policies among fathers but not among mothers. But neither the experience of having a daughter in general nor the proportion of daughters a man fathers affects support for gender-equity policies. & men/women, for fathers: Democrat/Republican, age of entry into fatherhood & Support for gender-equality policies, support for liberal policies among fathers (specification: sex of first child) & Specification: entropy balancing model (Table 2) uses 'has daughter' specification without controlling for no. of children
Inclusion of nonbiological offspring: yes
Post-treat: controls for partisan affiliation, ideology, gender-equality attitudes scale, hostile sexism scale
Other data concerns: study defines "fathers" and "mothers" as heterosexual individuals, small sample size \\ 
   \midrule
\cite{greenlee2020helping} & 2016 & US fathers (no more than 5 children) & Men whose first child is a girl, when compared to men whose first child was a boy, were (1) more likely to vote for Clinton on Election Day and (2) more likely to support a fictional congressional candidate when the candidate makes an appeal touting the importance of her candidacy for women and girls. Having a daughter as a first child did not influence mothers� voting patterns. &  & Pre-election Clinton preference (vs all other candidates), vote for Clinton in 2016, vote for Obama in 2012 (placebo), support for fictional female Congressional candidate & Specification: Models use 'has daughter' specification without controlling for no. of children
Inclusion of nonbiological offspring: yes
Post-treat: controls for partisan affiliation, ideology, support for gender equity
Other data concerns: small sample size \\ 
   \midrule
\cite{perales2018men} & 2001, 2005, 2008, 2011, 2015 & Australian adults & Among men, having a firstborn daughter is associated with a significantly larger increase in support for traditional gender-role attitudes than having a firstborn son. For women, there are no statistically significant differences. For female and male parents of firstborn daughters as well as for male parents of firstborn sons, the number of years after the birth of the firstborn child is not statistically related to gender-role attitudes. For female parents of firstborn sons, however, the model suggests a trend toward less traditional gender attitudes over time. & fathers/mothers & Gender-role attitudes (level of agreement with 7 statements) & Inclusion of nonbiological offspring: ambiguous \\ 
   \midrule
\cite{clayton2019daughters} & 2004--2012 & South African citizens & The effect of sex of first child on attitudes towards preferential hiring of women; views on abortion, gender equality, support for ANC is generally indistinguishable from zero except that respondents whose first child is a son are slightly more progressive. &  & Attitudes towards preferential hiring of women, views on abortion in two cases (low income/birth defect), gender equality battery, partisan support for the dominant political party (the center-left ANC) & Inclusion of nonbiological offspring: yes
Other data concerns: authors subset sample to respondents aged 45 and younger, biasing towards families with younger children \\ 
   \midrule
\cite{costa2019family} & 2007--2016 & US male Representatives & Cannot reject the null hypothesis that having a daughter has no effect on support for women�s issues among fathers or mothers in Congress. & Democrat/Republican & AAUW scores, cosponsorships (on bills with female sponsor vs male sponsor) & Specification: no measure of proportion of daughters, effect of no. of children may be nonlinear
Inclusion of nonbiological offspring: yes
Post-treat: controls for Representatives' partisan affiliation (and interaction of 'has daughter' with party ID), cosponsorship table controls for Republican x Republican bill sponsor interaction effect \\ 
   \midrule
\cite{borrell2019mighty} & 1991--2012 & UK parents & Results suggest that having daughters is associated with lower levels of support for traditional gender norms among men. For women, the association is ambiguous. Among men, the size of the coefficient is approximately halved when accounting for time-invariant unobserved heterogeneity. & men/women & Attitudes towards traditional gender norms (including traditional gender division of labor and within-suubject attitudinal shifts),  
for fathers: household gender division of labor & Specification: no measure of proportion of daughters
Inclusion of nonbiological offspring: yes
Other data concerns: authors restrict sample to respondents with children who are 21 years old or younger and still living in the household (biased towards younger families) \\ 
   \midrule
\cite{van2020papa} & 1997�1999 (US Congressmen), 1974 (French Congressmen), 2014 (European men), 2006 (American men) & US and French male legislators, US and French men & Among right-wing French male politicians, having one additional daughter leads to an 8 percentage points decrease in the probability of voting in favor of abortion law.  In contrast, there is no evidence of an impact of the presence of daughters on the left-wing congressmen�s vote. Among US Democratic MCs, the average propensity to vote left increases by more than 7 percentage points with each female child. The point estimate is negative and not significant for Republican congressmen. Among men in the US general population, one additional daughter is associated with more pro-abortion views for liberal respondents, while there are no significant effect for conservative respondents. & Democrat/Republican (for French legislators: Left/Right), liberal/conservative & French legislators DV: roll call votes on 1974 abortion law
US legislators DV: roll call votes on teen access to abortion
Citizens: attitudes toward abortion rights
US citizens: opposition to laws that prohibit abortion if they were in the situation of voting in Congress & Inclusion of nonbiological offspring: yes \\ 
   \midrule
\cite{pope2021father} & 1787 & Members of the Constitutional Convention & Each additional son is associated with a delegate being about 8 percent more likely to cast a pro-national ballot on any of the eight measures included in the index. In contrast, each additional daughter is associated with a 5.5 percent drop in probability of casting a pro-national ballot. &  & Convention votes that express delegate desires to strengthen or weaken the national government, whether delegates signed the Constitution (non-signers = anti-federalists) & Specification: effects of number of sons and number of daughters may not be linear
Other data concerns: small sample size \\ 
   \midrule
\cite{wesley2021effect} & 2016 & US adults & Having sons depresses the probability that individuals will identify as a feminist or strong feminist. But no effect of having daughters only or having both sons and daughters on feminist self-identity (no discernible differences in feminist self-identity for individuals with daughters only, both sons and daughters, and no children). Having only sons increases support for traditional gender roles for both men and women.   Having daughters does not increase the probability that men or women perceive it important to elect more women.  Women with sons only are less likely to perceive that it is important to elect women representatives, but this effect is not observed for men. & men/women & Feminist self-identification, attitudes toward women in elected office, attitudes toward traditional gender roles & Specification: compares any daughter(s)/any son(s) to no children, no measure of proportion of daughters, uses categorical specification without controlling for no. of children
Inclusion of nonbiological offspring: yes
Post-treat: controls for partisan identification, ideological self-identification, policy ideology scale, moral traditionalism scale, equality scale \\ 
   \bottomrule
\label{tab:lit}
\end{longtable}
\endgroup
